Anticipatory Collection of Metrics and Logs

1. A system comprising: a processor configured to: collect resource usage data regarding resource use within a computing system, the resource usage data being collected periodically, without running a diagnostic program, and before occurrence of a diagnosis event; start the diagnostic program to contemporaneously collect diagnosis data in response to the occurrence of the diagnosis event; convert collected resource usage data into a format used by the diagnostic program and timestamp the collected resource usage data to indicate that the collected resource usage data occurred prior to the occurrence of the diagnosis event; and provide the collected resource usage data to the diagnostic program executed after the occurrence of the diagnosis event so that the diagnostic program has an integrated view of resource usage data from before the occurrence of the diagnosis event with the diagnosis data collected by the diagnostic program in response to the occurrence of the diagnosis event to enable determination of a cause of the diagnosis event.

2. The system of claim 1 , wherein the collected resource usage data from before the occurrence of the diagnosis event and the diagnosis data collected by the diagnostic program in real time after the occurrence of the diagnosis event are processed by the diagnostic program to provide to an analyzer that analyzes the processed data to determine the cause of the diagnosis event.

3. The system of claim 1 , wherein collecting the resource usage data periodically and without running the diagnostic program consumes fewer resources than collecting the diagnosis data by running the diagnostic program.

4. The system of claim 1 , wherein the collected resource usage data includes data regarding use of one or more of processing and storage resources by the computing system.

5. The system of claim 1 , wherein the processor is further configured to store the collected resource usage data in a circular buffer having a predetermined amount of memory.

6. The system of claim 1 , wherein the processor is further configured to collect and store the resource usage data using less than or equal to a predetermined amount of resources allocated to the computing system.

7. The system of claim 1 , wherein the processor is further configured to: store at least some of the collected resource usage data in a compressed form, and provide the collected resource usage data in a decompressed form to the diagnostic program.

8. The system of claim 1 , wherein the collected resource usage data is used by an analyzer to generate a recommendation for reconfiguring the computing system or for reconfiguring one or more resources of the computing system.

9. The system of claim 1 , wherein the collected resource usage data is stored in a compressed format and wherein the processor is further configured to convert the collected resource usage data into the format used by the diagnostic program by expanding the collected resource usage data and populating the collected resource usage data into corresponding rows of a data table used by the diagnosis program for analysis of the diagnosis event.

10. A method comprising: collecting resource usage data regarding resource use within a computing system; collecting the resource usage data periodically, without running a diagnostic program, and before occurrence of an event to be diagnosed; starting the diagnostic program to contemporaneously collect diagnosis data in response to the occurrence of the event; converting collected resource usage data into a format used by the diagnostic program and timestamp the collected resource usage data to indicate that the collected resource usage data occurred prior to the occurrence of the event; and providing the collected resource usage data to the diagnostic program executed after the occurrence of the event so that the diagnostic program has an integrated view of resource usage data from before the occurrence of the event with the diagnosis data collected by the diagnostic program in response to the occurrence of the event to enable determination of a cause of the event.

11. The method of claim 10 , further comprising providing the collected resource usage data to the diagnostic program with timestamps indicating times when the data was collected.

12. The method of claim 10 , wherein the collected resource usage data from before the occurrence of the event and diagnosis data collected independently by the diagnostic program in real time after the occurrence of the event are processed by the diagnostic program to enable determination of the cause of the event.

13. The method of claim 10 , wherein collecting the resource usage data periodically and without running the diagnostic program consumes less resources than collecting diagnosis data by running the diagnostic program.

14. The method of claim 10 , further comprising collecting the resource usage data including data regarding use of one or more of processing and storage resources by the computing system.

15. The method of claim 10 , further comprising storing the collected resource usage data in a circular buffer having a predetermined amount of memory.

16. The method of claim 10 , further comprising collecting and storing the resource usage data using less than or equal to a predetermined amount of resources allocated to the computing system.

17. The method of claim 10 , further comprising: storing at least some of the collected resource usage data in a compressed form; and providing the collected resource usage data in a decompressed form to the diagnostic program.

18. The method of claim 10 , further comprising providing a recommendation for reconfiguring the computing system or for reconfiguring one or more resources of the computing system based on the collected resource usage data.

19. A system comprising: a processor configured to: collect resource usage data regarding use of one or more of processing and storage resources by a computing system, the resource usage data being collected periodically in a circular buffer, without running a diagnostic program, and before occurrence of an event to be diagnosed by the diagnostic program; start the diagnostic program to contemporaneously collect diagnosis data in response to the occurrence of the event; convert collected resource usage data into a format used by the diagnostic program and timestamp the collected resource usage data to indicate that the collected resource usage data occurred prior to the occurrence of the event; and provide the collected resource usage data, with the timestamp indicating times when the data was collected, to the diagnostic program when the diagnostic program is executed after the occurrence of the event so that the diagnostic program has an integrated view of the resource usage data, from before the occurrence of the event and the diagnosis data collected independently by the diagnostic program in real time after the occurrence of the event, to enable determination of a cause of the event.

20. The system of claim 19 , wherein the processor is further configured to collect and store the resource usage data using less than or equal to a predetermined amount of resources allocated to the computing system, and wherein collecting the resource usage data periodically and without running the diagnostic program consumes fewer resources than collecting diagnosis data by running the diagnostic program.